Real-time Mental Stress Detection Technique Using Neural Networks towards Wearable Health Monitor

2021 ◽  
Author(s):  
Nilava Mukherjee ◽  
Sumitra Mukhopadhyay ◽  
Rajarshi Gupta
Keyword(s):  
Feature Extraction ◽  
Real Time ◽  
Mental Stress ◽  
Detection Technique ◽  
Physiological Signals ◽  
Support Vector ◽  
Discrete Wavelet ◽  
Optimum Number ◽  
Stress Detection ◽  
Detection Techniques

Abstract Motivation: In recent times, mental stress detection using physiological signals have received widespread attention from the technology research community. Although many motivating research works have already been reported in this area, the evidence of hardware implementation is occasional. The main challenge in stress detection research is using optimum number of physiological signals, and real-time detection with low complexity algorithm. Objective: In this work, a real-time stress detection technique is presented which utilises only photoplethysmogram (PPG) signal to achieve improved accuracy over multi-signal-based mental stress detection techniques. Methodology: A short segment of 5s PPG signal was used for feature extraction using an autoencoder (AE), and features were minimized using recursive feature elimination (RFE) integrated with a multi-class support vector machine (SVM) classifier. Results: The proposed AE-RFE-SVM based mental stress detection technique was tested with WeSAD dataset to detect four-levels of mental state, viz., baseline, amusement, meditation and stress and to achieve an overall accuracy, F1 score and sensitivity of 99%, 0.99 and 98% respectively for 5s PPG data. The technique provided improved performance over discrete wavelet transformation (DWT) based feature extraction followed by classification with either of the five types of classifiers, viz., SVM, random forest (RF), k-nearest neighbour (k-NN), linear regression (LR) and decision tree (DT). The technique was translated into a quad-core-based standalone hardware (1.2 GHz, and 1 GB RAM). The resultant hardware prototype achieves a low latency (~0.4 s) and low memory requirement (~1.7 MB). Conclusion: The present technique can be extended to develop remote healthcare system using wearable sensors.

scholarly journals Anxiety Level Recognition for Virtual Reality Therapy System Using Physiological Signals

Electronics ◽  
2019 ◽  
Vol 8 (9) ◽  
pp. 1039 ◽  
Author(s):  
Justas Šalkevicius ◽  
Robertas Damaševičius ◽  
Rytis Maskeliunas ◽  
Ilona Laukienė
Keyword(s):  
Virtual Reality ◽  
Mental Stress ◽  
Cross Validation ◽  
Detection System ◽  
Anxiety Level ◽  
Physiological Signals ◽  
Support Vector ◽  
Svm Classifier ◽  
Stress Detection ◽  
The Individual

Virtual reality exposure therapy (VRET) can have a significant impact towards assessing and potentially treating various anxiety disorders. One of the main strengths of VRET systems is that they provide an opportunity for a psychologist to interact with virtual 3D environments and change therapy scenarios according to the individual patient’s needs. However, to do this efficiently the patient’s anxiety level should be tracked throughout the VRET session. Therefore, in order to fully use all advantages provided by the VRET system, a mental stress detection system is needed. The patient’s physiological signals can be collected with wearable biofeedback sensors. Signals like blood volume pressure (BVP), galvanic skin response (GSR), and skin temperature can be processed and used to train the anxiety level classification models. In this paper, we combine VRET with mental stress detection and highlight potential uses of this kind of VRET system. We discuss and present a framework for anxiety level recognition, which is a part of our developed cloud-based VRET system. Physiological signals of 30 participants were collected during VRET-based public speaking anxiety treatment sessions. The acquired data were used to train a four-level anxiety recognition model (where each level of ‘low’, ‘mild’, ‘moderate’, and ‘high’ refer to the levels of anxiety rather than to separate classes of the anxiety disorder). We achieved an 80.1% cross-subject accuracy (using leave-one-subject-out cross-validation) and 86.3% accuracy (using 10 × 10 fold cross-validation) with the signal fusion-based support vector machine (SVM) classifier.

Real-Time EEG Device for Epilepsy Detection Using Wavelet Transform and Support Vector Machine

2021 ◽  
pp. 104-135
Author(s):  
Sharad Sarjerao Jagtap ◽  
Rajesh Kumar M.
Keyword(s):  
Support Vector Machine ◽  
Feature Extraction ◽  
Wavelet Transform ◽  
Real Time ◽  
Low Cost ◽  
Support Vector ◽  
Svm Classifier ◽  
Discrete Wavelet ◽  
Feature Extraction Technique

This chapter gives an effective and efficient technique that can detect epilepsy in real time. It is low cost, low power, and real-time devices that can easily detect epilepsy. Along with EEG device, one can upgrade with GSM module to alert the doctors and parents of patients about its occurrence to prevent a sudden fall, which may cause injury and death. The accuracy of this EEG device depends on the quality of feature extraction technique and classification algorithm. In this chapter, support vector machine (SVM) is used as a classifier. Wavelet transform gives feature extraction, which helps to train data and to detect normal or seizure patients. Discrete wavelet transform (DWT) decomposes the signals into three decomposition levels. In this detection, mean, median, and non-linear parameter entropy were calculated for every sub-band as key parameters. The extracted features are then applied to SVM classifier for the classification. Better accuracy of classification is obtained using wavelet and SVM classifier.

scholarly journals Design and Implementation of Electromyography (EMG) based Real-Time Pattern Recognition model for Prosthetic hand Control

2021 ◽  
Author(s):  
Pasan Yashoda Jayaweera
Keyword(s):  
Feature Extraction ◽  
Real Time ◽  
Window Size ◽  
Machine Learning Algorithms ◽  
Support Vector ◽  
Extraction Techniques ◽  
Time Pattern ◽  
Feed Forward ◽  
Prosthetic Devices ◽  
Detection Techniques

Surface electromyogram (EMG) signals are a key component in myoelectric control systems utilized in modern prosthetic devices. Despite extensive study into EMG gesture detection techniques for hand and arm gestures, most prosthetic devices rely on direct control approaches that are often confined to single movements. The purpose of this paper is to investigate various feature extraction techniques and to compare various machine learning algorithms, window sizes to identify the most suitable algorithm, window size for real-time gesture recognition. For this purpose, a publicly available pre-labeled 2-channel EMG dataset was used as EMG signals. Feature sets for each window size were extracted using various feature extraction techniques and fed into support vector machines, k-nearest neighbors, ensemble learning, and feed-forward artificial neural network (ANN) classifiers. The feed-forward neural networks classifier was determined to be the best classifier based on its accuracies, sizes, and prediction delays for each window size. The maximum accuracy of the feed-forward ANN classifier was ≈87% with a 300-millisecond window size. the use of the majority voting technique was considered in terms of the number of votes and the window sizes.

Real Time Mental Stress Detection Through Breath Analysis

2020 ◽  
pp. 403-410
Author(s):  
Paulo Santos ◽  
Peter Roth ◽  
Jorge M. Fernandes ◽  
Viktor Fetter ◽  
Valentina Vassilenko
Keyword(s):  
Real Time ◽  
Mental Stress ◽  
Breath Analysis ◽  
Stress Detection

Soft computing based audio signal analysis for accident prediction

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Hima Bindu Valiveti ◽  
Anil Kumar B. ◽  
Lakshmi Chaitanya Duggineni ◽  
Swetha Namburu ◽  
Swaraja Kuraparthi
Keyword(s):  
Feature Extraction ◽  
Audio Signal ◽  
Support Vector ◽  
Classification Algorithms ◽  
Spectral Features ◽  
Road Accidents ◽  
Content Type ◽  
Detection Techniques ◽  
Car Crashes ◽  
High Level

Purpose Road accidents, an inadvertent mishap can be detected automatically and alerts sent instantly with the collaboration of image processing techniques and on-road video surveillance systems. However, to rely exclusively on visual information especially under adverse conditions like night times, dark areas and unfavourable weather conditions such as snowfall, rain, and fog which result in faint visibility lead to incertitude. The main goal of the proposed work is certainty of accident occurrence. Design/methodology/approach The authors of this work propose a method for detecting road accidents by analyzing audio signals to identify hazardous situations such as tire skidding and car crashes. The motive of this project is to build a simple and complete audio event detection system using signal feature extraction methods to improve its detection accuracy. The experimental analysis is carried out on a publicly available real time data-set consisting of audio samples like car crashes and tire skidding. The Temporal features of the recorded audio signal like Energy Volume Zero Crossing Rate 28ZCR2529 and the Spectral features like Spectral Centroid Spectral Spread Spectral Roll of factor Spectral Flux the Psychoacoustic features Energy Sub Bands ratio and Gammatonegram are computed. The extracted features are pre-processed and trained and tested using Support Vector Machine (SVM) and K-nearest neighborhood (KNN) classification algorithms for exact prediction of the accident occurrence for various SNR ranges. The combination of Gammatonegram with Temporal and Spectral features of the validates to be superior compared to the existing detection techniques. Findings Temporal, Spectral, Psychoacoustic features, gammetonegram of the recorded audio signal are extracted. A High level vector is generated based on centroid and the extracted features are classified with the help of machine learning algorithms like SVM, KNN and DT. The audio samples collected have varied SNR ranges and the accuracy of the classification algorithms is thoroughly tested. Practical implications Denoising of the audio samples for perfect feature extraction was a tedious chore. Originality/value The existing literature cites extraction of Temporal and Spectral features and then the application of classification algorithms. For perfect classification, the authors have chosen to construct a high level vector from all the four extracted Temporal, Spectral, Psycho acoustic and Gammetonegram features. The classification algorithms are employed on samples collected at varied SNR ranges.

scholarly journals Copy Move Forgery Detection Through Differential Excitation Component-Based Texture Features

2020 ◽  
Vol 12 (3) ◽  
pp. 27-44
Author(s):  
Gulivindala Suresh ◽  
Chanamallu Srinivasa Rao
Keyword(s):  
Feature Vector ◽  
Texture Feature ◽  
Texture Features ◽  
Support Vector ◽  
Discrete Wavelet ◽  
Detection Accuracy ◽  
Detection Techniques ◽  
Copy Move Forgery Detection ◽  
Occurrence Matrix ◽  
Differential Excitation

Copy-move forgery (CMF) is an established process to copy an image segment and pastes it within the same image to hide or duplicate a portion of the image. Several CMF detection techniques are available; however, better detection accuracy with low feature vector is always substantial. For this, differential excitation component (DEC) of Weber Law descriptor in combination with the gray level co-occurrence matrix (GLCM) approach of texture feature extraction for CMFD is proposed. GLCM Texture features are computed in four directions on DEC and this acts as a feature vector for support vector machine classifier. These texture features are more distinguishable and it is validated through other two proposed methods based on discrete wavelet transform-GLCM (DWT-GLCM) and GLCM. Experimentation is carried out on CoMoFoD and CASIA databases to validate the efficacy of proposed methods. Proposed methods exhibit resilience against many post-processing attacks. Comparative analysis with existing methods shows the superiority of the proposed method (DEC-GLCM) with regard to detection accuracy.

WATERMARK DETECTION USING DYNAMIC STOCHASTIC RESONANCE

2013 ◽  
Vol 12 (03) ◽  
pp. 1350010 ◽  
Author(s):  
RAJIB KUMAR JHA ◽  
APOORV CHATURVEDI ◽  
RAJLAXMI CHOUHAN
Keyword(s):  
Stochastic Resonance ◽  
Spread Spectrum ◽  
Internal Noise ◽  
External Noise ◽  
Detection Technique ◽  
Discrete Wavelet ◽  
Detection Techniques ◽  
Dynamic Stochastic Resonance ◽  
Watermark Embedding ◽  
Dynamic Stochastic

In this paper, a dynamic stochastic resonance (DSR) based watermark detection technique in discrete wavelet transform (DWT) domain is presented. Pseudo random bit sequence having certain seed value is considered as a watermark. Watermark embedding is done by embedding random bits in spread-spectrum fashion to the significant DWT coefficients. Watermark detection is quantitatively characterized by the value of correlation. The performance of watermark detection is improved by DSR which is an iterative process that utilizes the internal noise present in the image or external noise which is added during attacks. Even under various noise attacks, geometrical distortions, image enhancement and compression attacks, the DSR-based random bits detection is observed to give noteworthy improvement over existing watermark detection techniques. DSR-based technique is also found to give better detection performance when compared with the suprathreshold stochastic resonance-based detection technique.

scholarly journals Fault Diagnosis of a Rotor and Ball-Bearing System Using DWT Integrated with SVM, GRNN, and Visual Dot Patterns

Sensors ◽  
2019 ◽  
Vol 19 (21) ◽  
pp. 4806 ◽  
Author(s):  
Wen-Lin Chu ◽  
Chih-Jer Lin ◽  
Kai-Chun Kao
Keyword(s):  
Motor Control ◽  
Real Time ◽  
Ball Bearing ◽  
Warning System ◽  
Support Vector ◽  
Discrete Wavelet ◽  
General Regression Neural Networks ◽  
Experimental Process ◽  
Time Motor ◽  
General Regression

In this study, a set of methods for the inspection of a working motor in real time was proposed. The aim was to determine if ball-bearing operation is normal or abnormal and to conduct an inspection in real time. The system consists of motor control and measurement systems. The motor control system provides a set fixed speed, and the measurement system uses an accelerometer to measure the vibration, and the collected signal data are sent to a PC for analysis. This paper gives the details of the decomposition of vibration signals, using discrete wavelet transform (DWT) and computation of the features. It includes the classification of the features after analysis. Two major methods are used for the diagnosis of malfunction, the support vector machines (SVM) and general regression neural networks (GRNN). For visualization and to input the signals for visualization, they were input into a convolutional neural network (CNN) for further classification, as well as for the comparison of performance and results. Unique experimental processes were established with a particular hardware combination, and a comparison with commonly used methods was made. The results can be used for the design of a real-time motor that bears a diagnostic and malfunction warning system. This research establishes its own experimental process, according to the hardware combination and comparison of commonly used methods in research; a design for a real-time diagnosis of motor malfunction, as well as an early warning system, can be built thereupon.

Sign in / Sign up

Export Citation Format

Share Document